First public release.
rMASC() simulates the Multi-Attribute Search and Choice (MASC) model of
Gluth, Deakin and Rieskamp (2026) — sequential information search, Bayesian
belief updating, and choice.Sigma_belief argument enables the multivariate MASC-C belief update:
when the assumed correlation structure is non-diagonal, observing one
attribute updates beliefs about correlated attributes via a Kalman filter
("belief spread"). With a diagonal or NULL Sigma_belief the model reduces
exactly to the original univariate MASC update.Sigma_true argument generates stimuli with a specified correlation
structure (a matrix, or a single uniform off-diagonal correlation).hotelgluth2024 dataset from the hotel-choice experiment.